In [ ]:
import matplotlib.pyplot as plt
dataset_df_texto_sample['word2vec_similarity'].hist().plot()
plt.title('Similaridad word2vec histograma')
Out[ ]:
Text(0.5, 1.0, 'Similaridad word2vec histograma')
Fecha: 2023-09-08
Coder: Lenin G. Falconí
Objetivo:
Dataset
El dataset a utilizar es el resultado de combinar los registros obtenidos de la Policía en conjunción con los tipos penales definidos en SIAF con la fecha de corte de 25 de agosto de 2023 y los tipos penales definidos por la Comisión Estadística i.e. $\mathbb{D}^{PN\cup FGE\cup CE}_{1210342 \times 119 }$
import pandas as pd
from src.utils import conectar_sql
from sqlalchemy import text
conx = conectar_sql()
table_name = 'relatosPoliciaSiaf_20230830'
query = text(f"select * from reportes.`{table_name}`;")
dataset_df = pd.read_sql(query, conx)
dataset_df.shape
2023-11-24 14:31:42.447116: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: SSE4.1 SSE4.2 AVX AVX2 FMA To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
conectando con big data database.... <sqlalchemy.engine.base.Connection object at 0x7f66d5209bd0>
(1210342, 119)
dataset_df.columns.tolist()
['index_policia', 'NDD', 'RelatoPolicia', 'ANIO_REGISTRO_POLICIABD', 'index_siaf', 'Fecha_Registro', 'Hora_Registro', 'Fecha_Incidente', 'Hora_Incidente', 'Tentativa', 'Direccion', 'Presunto_Delito', 'Presunto Delito (Circunstancia Modificatoria)', 'Fiscalia', 'Fiscalia_Especializada', 'Ciudad', 'PROVINCIA', 'Canton', 'Parroquia', 'Tipo', 'NyF', 'numero_fiscalia', 'edificio', 'nombre_fiscal', 'Fuero', 'Etapa_procesal', 'Estado_Procesal', 'Fecha_estado_procesal', 'IMPULSOS_DILIGENCIAS', 'IMPULSOS', 'Ultima_accion', 'Fecha_Ultima_Accion', 'APREHENDIDO', 'SOSPECHOSO', 'PROCESADO', 'DETENIDO', 'IMPUTADO', 'numeroinf', 'Tipo_incidente', 'Fecha_IP_Inicio', 'Fecha_IP_Fin', 'Fecha_INS_Inicio', 'Fecha_INS_Fin', 'Tiempo_INS', 'RelatoFiscalia', 'PROVINCIA_INCIDENTE', 'CANTON_INCIDENTE', 'cantidadPalabras_RelatoPolicia', 'cantidadPalabras_RelatoFiscalia', 'RelatosIguales', 'Fecha_Registro_comision', 'Hora_Registro_comision', 'Fecha_Incidente_comision', 'Hora_Incidente_comision', 'Menor_comision', 'Mayor_comision', 'Tentativa_comision', 'Sector_comision', 'Modalidad_comision', 'Presunto_Delito_comision', 'Brigada_PJ_comision', 'Fiscalia_comision', 'Fiscalia_Especializada_comision', 'Ciudad_comision', 'Parroquia_comision', 'bar_descripcion_comision', 'Fuero_comision', 'Lugar_comision', 'Tipo_Delito_PJ_comision', 'Origen_comision', 'Tipo_comision', 'Bien_comision', 'Tipo_Involucrado_comision', 'edad_actual_comision', 'estatus_migratorio_comision', 'autodef_etnica_comision', 'FyNF_comision', 'desagregacion_comision', 'lugar_desagregacion_comision', 'modalidad_desagregacion_comision', 'cod_provincia_comision', 'pro_descripcion_comision', 'cod_parroquia_comision', 'cod_canton_comision', 'cod_barrio_comision', 'fisbrigas_comision', 'numero_fiscalia_comision', 'edificio_comision', 'mes_i_comision', 'ANIO_I_comision', 'desagregacion_1_comision', 'delitos_validados_comision', 'delitos_seguimiento_comision', 'PROVIN_comision', 'Cant_5_comision', 'Canton_comision', 'sexo_comision', 'Estado_comision', 'Desplegar_comision', 'Vehiculo_comision', 'genero_comision', 'Fecha_Nacimiento_comision', 'uni_codigo_comision', 'paises_comision', 'longitud_comision', 'latitud_comision', 'FASE_PROCESAL_ACTUAL_comision', 'PROXIMA_FASE_PROCESAL_comision', 'PAIS_INVOLUCRADO_comision', 'ART_comision', 'gen_secuencial_codigo_comision', 'Direccion_comision', 'relato_comision', 'relato_siaf_comision', 'cantidad_palabras_comision', 'score_delitos_seguimiento_comision', 'score_delitos_validados_comision', 'compare_comision', 'FechaActualizacion_comision']
dataset_df[['NDD', 'Presunto_Delito', 'Tipo_Delito_PJ_comision', 'RelatoPolicia', 'cantidadPalabras_RelatoPolicia', 'RelatoFiscalia', 'cantidadPalabras_RelatoFiscalia', 'delitos_seguimiento_comision']].sample(5)
| NDD | Presunto_Delito | Tipo_Delito_PJ_comision | RelatoPolicia | cantidadPalabras_RelatoPolicia | RelatoFiscalia | cantidadPalabras_RelatoFiscalia | delitos_seguimiento_comision | |
|---|---|---|---|---|---|---|---|---|
| 1123957 | 170101817042581 | ROBO | ROBO | ES EL CASO SEÑOR FISCAL QUE EL DÍA, HORA Y EN ... | 115 | ES EL CASO SEÑOR FISCAL QUE EL DÍA, HORA ... | 121 | ROBO A DOMICILIO |
| 766190 | 090101818042355 | ABUSO DE CONFIANZA | None | SEÑOR FISCAL MI PADRE QUE RESPONDE A LOS NOMBR... | 300 | Se adjunta denuncia escrita, con 4 anexos en ... | 10 | None |
| 644113 | 030301820120041 | ESTAFA | None | SEÑOR FISCAL EL DÍA DE AYER 30 DE DICIEMBRE DE... | 85 | SEÑOR FISCAL EL DÍA DE AYER 30 DE DICIEMBRE DE... | 89 | None |
| 553754 | 070901820020017 | ROBO | ROBO | ES EL CASO SÑEOR FISCAL QUE EL DIA DE AYER 03 ... | 108 | ES EL CASO SÑEOR FISCAL QUE EL DIA DE AYER 03 ... | 113 | ROBO A PERSONAS |
| 22498 | 240301815020098 | HURTO | None | 0 | DENUNCIA REGISTADA EN LA FISCLIA MEDIANTE OFIC... | 9 | None |
Se aplica las siguientes técnicas para calcular la similitud de los relatos
import nltk
from nltk.corpus import stopwords
nltk.download('stopwords')
stopWords = stopwords.words('spanish')
[nltk_data] Downloading package stopwords to [nltk_data] /home/falconiel/nltk_data... [nltk_data] Package stopwords is already up-to-date!
import transformers
import torch
import numpy as np
import pandas as pd
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
from gensim.models import Word2Vec
from transformers import BertTokenizer, BertModel, AutoTokenizer
from transformers import AutoModel
model_ckpt = "distilbert-base-multilingual-cased"
# from bert_score import score
def cosine_similarity_tfidf(text1, text2):
"""Calculates the cosine similarity between two documents using TF-IDF."""
vectorizer = TfidfVectorizer(stop_words=stopWords) # to adapt to spanish
vectors = vectorizer.fit_transform([text1, text2])
similarity = cosine_similarity(vectors[0], vectors[1])
return similarity.item()
def cosine_similarity_word2vec(text1, text2):
"""Calculates the cosine similarity between two documents using word2vec."""
sentences = [text1.split(), text2.split()]
model = Word2Vec(sentences, min_count=1)
embeddings1 = model.wv[text1.split()].mean(axis=0).reshape(1,-1)
embeddings2 = model.wv[text2.split()].mean(axis=0).reshape(1,-1)
# similarity = np.dot(embeddings1, embeddings2) / (np.linalg.norm(embeddings1) * np.linalg.norm(embeddings2))
similarity = cosine_similarity(embeddings1, embeddings2)
return similarity.item()
def cosine_similarity_bert(text1, text2):
"""Calculates the cosine similarity between two documents using BERT."""
tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
model = BertModel.from_pretrained('bert-base-uncased')
encoded_input1 = tokenizer(text1, return_tensors='pt')
encoded_input2 = tokenizer(text2, return_tensors='pt')
output1 = model(**encoded_input1)
output2 = model(**encoded_input2)
embeddings1 = output1.last_hidden_state.detach().numpy()
embeddings2 = output2.last_hidden_state.detach().numpy()
# print(embeddings1)
# similarity = cosine_similarity(embeddings1, embeddings2)
similarity_score = cosine_similarity(embeddings1.mean(axis=1), embeddings2.mean(axis=1))
return similarity_score.item()
def cosine_similarity_bert_model(text1, text2, model=model_ckpt):
"""Calculates the cosine similarity between two documents using bert but
specifying the model to be used
"""
tokenizer = AutoTokenizer.from_pretrained(model)
bert_model = AutoModel.from_pretrained(model)
# Tokeniza el texto y trunca si supera los 512 tokens
# inputs = tokenizer.encode(text, truncation=True, max_length=512)
encoded_input1 = tokenizer(text1, truncation=True, max_length=256, return_tensors='pt')
encoded_input2 = tokenizer(text2, truncation=True, max_length=256, return_tensors='pt')
output1 = bert_model(**encoded_input1)
output2 = bert_model(**encoded_input2)
embeddings1 = output1.last_hidden_state.detach().numpy()
embeddings2 = output2.last_hidden_state.detach().numpy()
# print(embeddings1.shape, embeddings2.shape)
similarity_score = cosine_similarity(embeddings1.mean(axis=1), embeddings2.mean(axis=1))
# similarity_score = cosine_similarity(embeddings1[0], embeddings2[0])
# print(similarity_score.shape)
return similarity_score.item()
sentences = ["El perro bebió té inglés.", "Mary tenía un libro de inglés."]
text1 = sentences[0]
text2 = sentences[1]
similarity_tfidf = cosine_similarity_tfidf(text1, text2)
similarity_word2vec = cosine_similarity_word2vec(text1, text2)
similarity_bert = cosine_similarity_bert(text1, text2)
similarity_distil_bert = cosine_similarity_bert_model(text1, text2, model=model_ckpt)
print("The cosine similarity between the two documents using TF-IDF is:", similarity_tfidf)
print("The cosine similarity between the two documents using word2vec is:", similarity_word2vec)
print("The cosine similarity between the two documents using BERT is:", similarity_bert)
print("The cosine similarity between the two documents using DISTIL-BERT is:", similarity_distil_bert)
Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertModel: ['cls.predictions.transform.LayerNorm.weight', 'cls.predictions.decoder.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.dense.bias', 'cls.seq_relationship.bias', 'cls.predictions.bias', 'cls.predictions.transform.LayerNorm.bias', 'cls.predictions.transform.dense.weight'] - This IS expected if you are initializing BertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing BertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
The cosine similarity between the two documents using TF-IDF is: 0.17077611319011649 The cosine similarity between the two documents using word2vec is: 0.11319118738174438 The cosine similarity between the two documents using BERT is: 0.8455851078033447 The cosine similarity between the two documents using DISTIL-BERT is: 0.6644577980041504
A fin de que los relatos se procesen de manera similar se aplicará el formateo del texto que se usa para la predicción desde el modelo de machine learning
from src.utils import format_crimestory
format_crimestory(dataf=dataset_df,
relato_label='RelatoPolicia')
format_crimestory(dataf=dataset_df,
relato_label='RelatoFiscalia')
dataset_df.RelatoFiscalia.isna().sum(), dataset_df.RelatoPolicia.isna().sum()
(0, 0)
dataset_df.cantidadPalabras_RelatoFiscalia.min(), dataset_df.cantidadPalabras_RelatoPolicia.min()
(0, 0)
No se puede realizar la comparacion en los casos en que no existe texto de policita o de fiscalia por ende se debe formar un nuevo dataset con los registros que sí tienen texto. Además por problemas de limitación de memoria se debe calcular una muestra representativa de los datos. Para esto se obtiene la siguiente tamaño de muestra con las condiciones indicadas y la ecuación:
$n =\frac{z^2p(1-p) }{\epsilon^2N+z^2p(1-p)}N $
Sample size: 46947
This means 46947 or more measurements/surveys are needed to have a confidence level of 99.999% that the real value is within ±1% of the measured/surveyed value.
# dataset_df_sample = dataset_df.sample(int(0.7*dataset_df.shape[0]))
# dataset_df_sample.shape
# del dataset_df_sample
# reduced_columns = ['NDD', 'RelatoPolicia', 'cantidadPalabras_RelatoPolicia', 'RelatoFiscalia', 'cantidadPalabras_RelatoFiscalia']
dataset_df_texto = dataset_df[(dataset_df.cantidadPalabras_RelatoPolicia>=50)&(dataset_df.cantidadPalabras_RelatoFiscalia>=50)]
dataset_df_texto.shape
(785513, 119)
Sample size: 45983
This means 45983 or more measurements/surveys are needed to have a confidence level of 99.999% that the real value is within ±1% of the measured/surveyed value.
Sample size: 16297
This means 16297 or more measurements/surveys are needed to have a confidence level of 99% that the real value is within ±1% of the measured/surveyed value.
Sample size: 6748
This means 6748 or more measurements/surveys are needed to have a confidence level of 90% that the real value is within ±1% of the measured/surveyed value.
https://www.calculator.net/sample-size-calculator.html?type=1&cl=90&ci=1&pp=50&ps=785513&x=Calculate
del dataset_df
SAMPLE_SIZE = 6000
dataset_df_texto_sample = dataset_df_texto[['NDD',
'RelatoFiscalia',
'cantidadPalabras_RelatoFiscalia',
'RelatoPolicia',
'cantidadPalabras_RelatoPolicia',
'Presunto_Delito',
'Tipo_Delito_PJ_comision',
'delitos_seguimiento_comision']].sample(SAMPLE_SIZE)
dataset_df_texto_sample.shape
(6000, 8)
from datasets import Dataset
dataset_df_texto_sample_hf = Dataset.from_pandas(dataset_df_texto_sample)
# dataset_df_texto['cosine_similarity'] = dataset_df_texto.apply(lambda x: cosine_similarity(x.RelatoPolicia, x.RelatoFiscalia), axis=1)
# dataset_df_texto_sample['word2vec_similarity'] = dataset_df_texto_sample.apply(lambda x: cosine_similarity_word2vec(x.RelatoPolicia, x.RelatoFiscalia), axis=1)
# dataset_df_texto_sample['cosine_similarity_bert_model'] = dataset_df_texto_sample.apply(lambda x: cosine_similarity_bert_model(x.RelatoPolicia, x.RelatoFiscalia, model=model_ckpt), axis=1)
dataset_df_texto_sample_hf = dataset_df_texto_sample_hf.map(lambda example: {'word2vec_similarity': cosine_similarity_word2vec(example['RelatoPolicia'], example['RelatoFiscalia'])})
0%| | 0/6000 [00:00<?, ?ex/s]
dataset_df_texto_sample_hf = dataset_df_texto_sample_hf.map(lambda example:{'cosine_similarity_bert_model': cosine_similarity_bert_model(text1=example['RelatoPolicia'],
text2=example['RelatoFiscalia'],
model=model_ckpt)})
0%| | 0/6000 [00:00<?, ?ex/s]
Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of the model checkpoint at distilbert-base-multilingual-cased were not used when initializing DistilBertModel: ['vocab_layer_norm.weight', 'vocab_projector.weight', 'vocab_layer_norm.bias', 'vocab_projector.bias', 'vocab_transform.weight', 'vocab_transform.bias'] - This IS expected if you are initializing DistilBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing DistilBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
Convert the dataset back to a pandas dataframe
del dataset_df_texto_sample
dataset_df_texto_sample = dataset_df_texto_sample_hf.to_pandas()
dataset_df_texto_sample.word2vec_similarity.describe()
count 6000.000000 mean 0.957418 std 0.073233 min 0.310206 25% 0.961602 50% 0.975714 75% 0.984985 max 1.000000 Name: word2vec_similarity, dtype: float64
import matplotlib.pyplot as plt
dataset_df_texto_sample['word2vec_similarity'].hist().plot()
plt.title('Similaridad word2vec histograma')
Text(0.5, 1.0, 'Similaridad word2vec histograma')
dataset_df_texto_sample['cosine_similarity_bert_model'].describe()
count 6000.000000 mean 0.987899 std 0.023865 min 0.780015 25% 0.990691 50% 0.995396 75% 0.997255 max 1.000000 Name: cosine_similarity_bert_model, dtype: float64
dataset_df_texto_sample['cosine_similarity_bert_model'].hist().plot()
plt.title('Similitud Bert histograma')
Text(0.5, 1.0, 'Similitud Bert histograma')
Se busca revisar si la predicción realizada sobre el dataset con el relato de policia es diferente de la realizada con el relato de fiscalía
from src.utils import load_text_classification_model
PATH_MODEL_SEGUIMIENTOS = '/home/falconiel/ML_Models/robbery_tf20221113'
PATH_MODEL_VALIDADOS = '/home/falconiel/ML_Models/robbery_tf20230213'
model_ckpt = "distilbert-base-multilingual-cased"
SEQ_LEN = 300
THRESHOLD_WORDS_QTY = 50
modelo_delitos_seguimiento = load_text_classification_model(path2model=PATH_MODEL_SEGUIMIENTOS,
seq_len=SEQ_LEN,
threshold_words_qty=THRESHOLD_WORDS_QTY)
2023-11-25 15:36:36.086702: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: SSE4.1 SSE4.2 AVX AVX2 FMA To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags. 2023-11-25 15:36:36.114535: I tensorflow/core/common_runtime/process_util.cc:146] Creating new thread pool with default inter op setting: 2. Tune using inter_op_parallelism_threads for best performance. All model checkpoint layers were used when initializing TFDistilBertForSequenceClassification. All the layers of TFDistilBertForSequenceClassification were initialized from the model checkpoint at /home/falconiel/ML_Models/robbery_tf20221113. If your task is similar to the task the model of the checkpoint was trained on, you can already use TFDistilBertForSequenceClassification for predictions without further training.
from src.utils import predict_text_class_tqdm
predict_text_class_tqdm(dataf=dataset_df_texto_sample,
model=modelo_delitos_seguimiento,
label_relato='RelatoPolicia',
label_name='delitosSeguimientoPolicia',
words_qty_label='cantidadPalabras_RelatoPolicia',
threshold_words_qty=THRESHOLD_WORDS_QTY)
100%|██████████| 6000/6000 [24:48<00:00, 4.03it/s]
from src.utils import predict_text_class_tqdm
predict_text_class_tqdm(dataf=dataset_df_texto_sample,
model=modelo_delitos_seguimiento,
label_relato='RelatoFiscalia',
label_name='delitosSeguimientoFiscalia',
words_qty_label='cantidadPalabras_RelatoFiscalia',
threshold_words_qty=THRESHOLD_WORDS_QTY)
100%|██████████| 6000/6000 [26:42<00:00, 3.74it/s]
# Create a figure with 2 subplots (vertically stacked)
fig, axs = plt.subplots(2, 1)
# Plot the value counts of the first column in the first subplot
dataset_df_texto_sample['delitosSeguimientoPolicia'].value_counts().plot.barh(ax=axs[0])
axs[0].set_title('Predicciones sobre Relatos Policía')
# Plot the value counts of the second column in the second subplot
dataset_df_texto_sample['delitosSeguimientoFiscalia'].value_counts().plot.barh(ax=axs[1])
axs[1].set_title('Predicciones sobre Relatos Fiscalía')
# Display the plot
plt.tight_layout()
plt.show()
print(f"predicciones sobre relato fiscalia en el dataset")
dataset_df_texto_sample.delitosSeguimientoFiscalia.value_counts()
predicciones sobre relato fiscalia en el dataset
ROBO A PERSONAS 2633 ROBO A DOMICILIO 1019 ROBO A UNIDADES ECONOMICAS 774 ROBO DE BIENES, ACCESORIOS Y AUTOPARTES DE VEHICULOS 690 ROBO DE MOTOS 514 ROBO DE CARROS 370 Name: delitosSeguimientoFiscalia, dtype: int64
print(f"predicciones sobre relato fiscalia en el dataset")
dataset_df_texto_sample.delitosSeguimientoPolicia.value_counts()
predicciones sobre relato fiscalia en el dataset
ROBO A PERSONAS 2650 ROBO A DOMICILIO 1027 ROBO A UNIDADES ECONOMICAS 778 ROBO DE BIENES, ACCESORIOS Y AUTOPARTES DE VEHICULOS 665 ROBO DE MOTOS 509 ROBO DE CARROS 371 Name: delitosSeguimientoPolicia, dtype: int64
dataset_df_texto_sample[['NDD', 'RelatoPolicia', 'RelatoFiscalia', 'cantidadPalabras_RelatoPolicia', 'cantidadPalabras_RelatoFiscalia', 'delitosSeguimientoPolicia', 'delitosSeguimientoFiscalia']].sample(5)
| NDD | RelatoPolicia | RelatoFiscalia | cantidadPalabras_RelatoPolicia | cantidadPalabras_RelatoFiscalia | delitosSeguimientoPolicia | delitosSeguimientoFiscalia | |
|---|---|---|---|---|---|---|---|
| 151 | 121201822100027 | que el dí a 16 de octubre del 2022 a las 07h00... | que el día 16 de octubre del 2022 a las 07h00 ... | 182 | 190 | ROBO DE MOTOS | ROBO DE MOTOS |
| 5355 | 061001815110025 | es el caso señor fiscal que soy policia nacion... | es el caso señor fiscal que soy policia nacion... | 172 | 182 | ROBO A PERSONAS | ROBO DE CARROS |
| 1203 | 170101814113001 | es el caso señor fiscal que me encontraba en l... | es el caso señor fiscal que me encontraba en l... | 80 | 84 | ROBO A PERSONAS | ROBO A PERSONAS |
| 2914 | 090101816044376 | señor fiscal el día 22 de abril del 2016 aprox... | señor fiscal el día 22 de abril del 2016 aprox... | 91 | 93 | ROBO A UNIDADES ECONOMICAS | ROBO A UNIDADES ECONOMICAS |
| 726 | 170101816062161 | es el caso señor fiscal que el día 12 de junio... | es el caso señor fiscal que el día 12 de junio... | 217 | 229 | ROBO A DOMICILIO | ROBO A DOMICILIO |
dataset_df_texto_sample.delitos_seguimiento_comision.isna().sum(), dataset_df_texto_sample.shape
(2193, (6000, 15))
Para realizar la evaluación de las predicciones podemos considerar delitosSeguimientoPolicia y delitosSeguimientoFiscalia como $\hat{y}_{policia} $ y $\hat{y}_{fge} $ donde la variable delitos_seguimiento_comision se consiera el Ground Truth. Para esto dado que $\exists $ filas $r_i$ donde delitos_seguimiento_comision no tienen un valor definido, se procede a eliminar estas filas para la evaluacón a fin de realizar una comparación estable. A fin de no disponer de cortes en la ejecución del programa se recomienda salvar en disco la información existente
# dataset_df_texto_sample.to_csv('data/processed/resultadosDatasetSimilitudPrediccionesTextoPoliciaFGE.csv', index=False)
dataset_df_texto_sample_2eval = dataset_df_texto_sample[-dataset_df_texto_sample.delitos_seguimiento_comision.isna()]
dataset_df_texto_sample.shape, dataset_df_texto_sample_2eval.shape
((6000, 15), (3807, 15))
from sklearn.metrics import classification_report
print("Reporte de Clasificación de predicciones sobre relato Fiscalia")
print(classification_report(y_true=dataset_df_texto_sample_2eval.delitos_seguimiento_comision,
y_pred=dataset_df_texto_sample_2eval.delitosSeguimientoFiscalia))
Reporte de Clasificación de predicciones sobre relato Fiscalia
precision recall f1-score support
ROBO A DOMICILIO 0.80 0.92 0.86 564
ROBO A PERSONAS 0.90 0.95 0.93 1458
ROBO A UNIDADES ECONOMICAS 0.61 0.88 0.72 245
ROBO DE BIENES, ACCESORIOS Y AUTOPARTES DE VEHICULOS 0.85 0.95 0.89 493
ROBO DE CARROS 0.93 0.93 0.93 312
ROBO DE MOTOS 0.98 0.95 0.96 415
SIN INFORMACION 0.00 0.00 0.00 320
accuracy 0.86 3807
macro avg 0.72 0.80 0.76 3807
weighted avg 0.80 0.86 0.83 3807
/home/falconiel/anaconda3/envs/mlenv/lib/python3.10/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result)) /home/falconiel/anaconda3/envs/mlenv/lib/python3.10/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result)) /home/falconiel/anaconda3/envs/mlenv/lib/python3.10/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
from sklearn.metrics import classification_report
print("Reporte de Clasificación de predicciones sobre relato Policía")
print(classification_report(y_true=dataset_df_texto_sample_2eval.delitos_seguimiento_comision,
y_pred=dataset_df_texto_sample_2eval.delitosSeguimientoPolicia))
Reporte de Clasificación de predicciones sobre relato Policía
precision recall f1-score support
ROBO A DOMICILIO 0.80 0.93 0.86 564
ROBO A PERSONAS 0.90 0.95 0.93 1458
ROBO A UNIDADES ECONOMICAS 0.58 0.88 0.70 245
ROBO DE BIENES, ACCESORIOS Y AUTOPARTES DE VEHICULOS 0.86 0.93 0.89 493
ROBO DE CARROS 0.89 0.90 0.90 312
ROBO DE MOTOS 0.97 0.93 0.95 415
SIN INFORMACION 0.00 0.00 0.00 320
accuracy 0.85 3807
macro avg 0.72 0.79 0.75 3807
weighted avg 0.79 0.85 0.82 3807
/home/falconiel/anaconda3/envs/mlenv/lib/python3.10/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result)) /home/falconiel/anaconda3/envs/mlenv/lib/python3.10/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result)) /home/falconiel/anaconda3/envs/mlenv/lib/python3.10/site-packages/sklearn/metrics/_classification.py:1334: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
El presente estudio dispone como amaneza a la validez la restricción derivada de los recursos de hardware disponibles (i.e. tamaño de RAM y GPUs) de donde a fin de poder presentar un resultado computable se procedió a obtener una muestra aleatoria con un nivel de confianza del 90% con un error esperado del 1% sobre 785513 registros, lo que recomienda usar un tamaño de muestra de 6748. Sin embargo, el dataset se redondeó al entero inferior más próximo que sería 6000. Este cambio afecta en el error esperado que sería en realidad del 1.1%
Para la comparación de textos de policía $text_{policia}$ y de fiscalía $text_{fge}$ se utilizaron dos métricas de similitud. La una está basada en word2vec y la otra en transformadores utilizando concretamente distil_bert que dispone de entrenamiento en lenguaje español. De los histogramas presentados se observa que en promedio ambos textos se consideran similares con una media de 0.97 entre ambos valores. También se observa que el 75% de los datos supera el 0.95 de similitud en dos técnicas distintas
Adicional a esto se realiza la comparación de predicciones derivadas de ambos relatos. Para esto se descartaron textos inferiores a 50 palabras y las filas en donde la comisión no dispone de valor registrado. Puede observarse que las variaciones relativas entre el relato de policía y el de fiscalía son mínimas. En promedio se podría hablar de una diferencia global entre ambos del 0.16%
Con respecto a las métricas de evaluación del clasificador se observa un rendimiento general de 0.86 para el relato de Fiscalía y 0.85 para el relato de Policía. En donde la diferencia menor se puede explicar porque el modelo está entrenado en relato de fiscalía. Sin emargo, se puede observar que la inferencia y capacidad de generalización del modelo no está supedidata al origen del relato por cuanto las expresiones idiomáticas más o menos serán las mismas que tengan relación con la categoría del delito.
De lo anterior se confirmaría que la manera de mejorar al modelo no está en entrenar con datos de policia o de fiscalía sino en entrenar con la maor cantidad de datos posibles lo que implica un mayor consumo de hardware
valores_fge = dataset_df_texto_sample.delitosSeguimientoFiscalia.value_counts().values
valores_policia = dataset_df_texto_sample.delitosSeguimientoPolicia.value_counts().values
valores_fge.sum(), valores_policia.sum()
(6000, 6000)
# resultado de variación relativa en % por categoría
resp = np.abs(valores_fge-valores_policia)/valores_fge.sum()*100
resp
array([0.28333333, 0.13333333, 0.06666667, 0.41666667, 0.08333333,
0.01666667])
resp.mean()
0.16666666666666666